Early detection of cognitive decline in Alzheimer's disease using eye tracking

Abstract

Background: Patients with Alzheimer's disease (AD) are known to exhibit visuospatial processing impairment, as reflected in eye movements from the early stages of the disease. We investigated whether the pattern of gaze exploration during visual tasks could be useful for detecting cognitive decline at the earliest stage.

Methods: Sixteen AD patients (age: 79.1 ± 7.9 years, Mini Mental State Examination [MMSE] score: 17.7 ± 5.3, mean ± standard deviation) and 16 control subjects (age: 79.4 ± 4.6, MMSE score: 26.9 ± 2.4) participated. In the visual memory task, subjects memorized presented line drawings for later recall. In the visual search tasks, they searched for a target Landolt ring of specific orientation (serial search task) or color (pop-out task) embedded among arrays of distractors. Using video-oculography, saccade parameters, patterns of gaze exploration, and pupil size change during task performance were recorded and compared between AD and control subjects.

Results: In the visual memory task, the number of informative regions of interest (ROIs) fixated was significantly reduced in AD patients compared to control subjects. In the visual search task, AD patients took a significantly longer time and more saccades to detect the target in the serial but not in pop-out search. In both tasks, there was no significant difference in the saccade frequency and amplitude between groups. On-task pupil modulation during the serial search task was decreased in AD. The number of ROIs fixated in the visual memory task and search time and saccade numbers in the serial search task differentiated both groups of subjects with high sensitivity, whereas saccade parameters of pupil size modulation were effective in confirming normal cognition from cognitive decline with high specificity.

Discussion: Reduced fixation on informative ROIs reflected impaired attentional allocation. Increased search time and saccade numbers in the visual search task indicated inefficient visual processing. Decreased on-task pupil size during visual search suggested decreased pupil modulation with cognitive load in AD patients, reflecting impaired function of the locus coeruleus. When patients perform the combination of these tasks to visualize multiple aspects of visuospatial processing, cognitive decline can be detected at an early stage with high sensitivity and specificity and its progression be evaluated.

Keywords: Alzheimer’s disease; cognitive decline; eye tracking; fixation; pupil; saccade.

FIGURE 1

(A) The four images presented in the visual memory task. Blue circles, which are not visible to the subjects, indicate the informative regions of interest (ROIs) that characterize the image. We determined the locations of these informative regions based on previous studies (Mosimann et al., 2004; Matsumoto et al., 2011b). The radius of these ROIs was 1.9 deg in visual angle. (B) The eye-tracking system. A 17-inch computer display is placed 50 cm in front of the subject’s eyes. The images are presented on the display, and the subject’s eye movements (fixations and saccades) while looking at the image are recorded by the video-oculography camera (Eyelink 1000^®). The scanned area is defined by the convex envelope of all fixation points. (C) The images used in the visual search tasks. In the serial search task, the target Landolt ring is placed upwards and others are placed downwards. In the pop-out task, the target ring is red and the others are black. In both tasks, the number of rings was 4 or 48. (D) An example of time distribution in 10 s while a subject is looking at an image. The gray bars indicate the fixation periods, and the white bars indicate other times (saccades, blinks, face displacements, and so on) that indicate failure of fixation. The red rectangle indicates the 5 s that include the longest total time duration of fixations. (E) An example of a pupil size plot during the visual search task. Pupil size data (blue line) cannot be recorded during blinks (arrow). The red line is the upper envelope of pupil size data, which is not influenced by saccades.

FIGURE 2

(A) The result of image drawing of the visual memory task. Examples of images drawn by AD patients and control subjects are shown. The drawing scores tended to be lower in AD patients. (B–G) Comparison of saccade parameters in the visual memory task between AD patients and control subjects. By repeated measures two-way ANOVA, the difference between these groups was significant only in panel (G) the number of focused ROIs (see the text). (H) The ROC analysis of the number of focused ROIs, averaged for images 1-4. When the cut-off was set to 3.125, the sensitivity was 1.0 and the specificity was 0.538. *p < 0.05.

FIGURE 3

Comparisons of saccade parameters in the visual search tasks. T-test shows that AD patients required significantly longer search time and more saccades than control subjects, mainly in the serial search task. *p < 0.05.

FIGURE 4

The ROC analysis of search time and number of saccades in the visual search task (serial search).

FIGURE 5

(A) The change of pupil size ratio during the serial search task as the task progressed. The red and blue lines indicate the average data of AD patients and control subjects, respectively, and the width of the color bars around the red or blue lines indicate their standard deviations. The pupil dilatation of AD patients tended to be less than that of control subjects, especially in the 48 ring task. (B) Comparison of pupil size ratio averaged across all 10 images (4 rings or 48 rings) between AD and control groups, which shows that pupil dilatation of AD patients is significantly less than that of control subjects in the 48 ring task, by t-test. The correlation between the MMSE score and the pupil size ratio was significant only in the 48 ring task. (C) The ROC analysis of the pupil size ratio in the 48 ring task. When the cut-off value is set to 0.962, the sensitivity was 0.5 and the specificity was 1.0. *p < 0.05.